TensorComprehensions
TensorComprehensions copied to clipboard
A domain specific language to express machine learning workloads.
Hi everyone, First of all, let me thank you for putting so much effort into this accessible, versatile library. As a mathematician working at the intersection between shape analysis and...
I apologize if this question is not relevant to this forum. Does this framework support MPI? Thanks,
Tensor Comprehensions Github Issues Guidelines ---------------------------------------------- If you have a feature request or a bug report (build issue), please open an issue on Github and fill the template below so...
#138 only adds a skeleton for the CPU test and smoke checks compilation. Make this into a real test.
We need some way to express padding of input tensors. The current thought is a primitive pad(expr, expr), which returns the first argument *if it does no out-of-bounds reads*, and...
one of our users reported on slack channel that they were trying to translate the following Numpy code ``` def A_matvec_batch(A, X): n, m = X.shape Y = np.zeros((n, m))...
There is a big interest in having support for RNN loops in TC. Creating this master task to discuss further and track progress
do we not allow people to use same variable twice? also repro is available in #61 ``` FAILS: def softmax(float(N, D) I) -> (O, tmp) { tmp(n) max=! I(n, d)...
right now, everything is either an input or output. Reason being TC does not do any allocation by itself. We should find a better way to handle this cc Albert...
We currently support computed expressions in the indices on the RHS. We should plumb through computed expressions on the LHS too, e.g: hist(im(i)) +=! 1